NExT-QA introduces a multi-task VideoQA benchmark focusing on causal and temporal reasoning, revealing models' weak performance in deep understanding.

Junbin Xiao, Xindi Shang, Angela Yao et al.

VPN++ employs dual-level knowledge distillation to fuse RGB videos and 3D poses, achieving fast, robust activity recognition without pose input at inference.

Srijan Das, Rui Dai, Di Yang et al.

TAT-QA dataset combines tables and text; TAGOP model improves F1 score to 58.0%.

Fengbin Zhu, Wenqiang Lei, Youcheng Huang et al.

Unified inference learning methods (Hebbian, Fenchel BP) as recursive bilevel optimization, showing their relation to standard backpropagation.

Christopher Zach

Informed Equation Learning (iEQL) integrates expert knowledge and robust training strategies to support singular functions like log and division, achieving high-accuracy, interpretable models.

Matthias Werner, Andrej Junginger, Philipp Hennig et al.

NSA (Negative Selection Algorithm) excels in nonlinear high-dimensional anomaly detection, outperforming traditional models in accuracy and speed.

Kishor Datta Gupta, Dipankar Dasgupta

Proposes PDA framework using causal intervention to remove bad bias effects and introduce desired bias, improving recommendation accuracy.

Yang Zhang, Fuli Feng, Xiangnan He et al.

DCT-NeRF learns trajectory fields for stable dynamic novel view synthesis.

Chaoyang Wang, Ben Eckart, Simon Lucey et al.

Kernel Thinning compresses n points to √n with provably comparable MMD error, outperforming i.i.d. sampling.

Raaz Dwivedi, Lester Mackey

Kleister datasets challenge existing KIE models; best model achieved 81.77% and 83.57% F1 on NDA and Charity datasets.

Tomasz Stanisławek, Filip Graliński, Anna Wróblewska et al.

Proposes spectral descriptor-based functional mapping and manifold regularization for unpaired heterogeneous multimodal data analysis.

Maysam Behmanesh, Peyman Adibi, Jocelyn Chanussot et al.

Diffusion models outperform GANs in image synthesis, achieving FID scores of 2.97, 4.59, and 7.72 on ImageNet 128×128, 256×256, and 512×512 respectively, with efficient sampling and classifier guidance.

Prafulla Dhariwal, Alex Nichol

Survey on deep learning dialogue systems, covering model types, system types, evaluation methods, and emerging trends.

Jinjie Ni, Tom Young, Vlad Pandelea et al.

Inter-GPS integrates formal language and symbolic reasoning, constructing Geometry3K, significantly improving geometric problem solving accuracy.

Pan Lu, Ran Gong, Shibiao Jiang et al.

D2S system integrates query-based long-form QA and dense vector retrieval for automatic scientific paper-to-slide generation.

Edward Sun, Yufang Hou, Dakuo Wang et al.

FlingBot employs high-speed dynamic flinging for cloth unfolding, achieving over 80% coverage within 3 actions, outperforming static baselines.

Huy Ha, Shuran Song

DExperts combines expert and anti-expert models to control text generation at decoding time, significantly reducing GPT-3 toxicity.

Alisa Liu, Maarten Sap, Ximing Lu et al.

Using metric entropy, proved RNNs can optimally learn stable linear systems, establishing quantitative bounds on system realization.

Clemens Hutter, Recep Gül, Helmut Bölcskei

DiffSinger synthesizes high-quality singing via shallow diffusion, achieving a 0.11 MOS improvement.

Jinglin Liu, Chengxi Li, Yi Ren et al.

This study reveals that latent space similarities in ProtoPNet can be misleading under noise, with experiments showing JPEG compression and adversarial perturbations cause interpretability failures.

Adrian Hoffmann, Claudio Fanconi, Rahul Rade et al.